Researchers at the University of California, Davis have developed a computer-implemented method for accurately classifying congenital heart defects in newborns using pulse oximetry and machine learning.

Researchers at the University of California, Davis have developed an atmospheric pressure ionization source that employs an ordered array of micro-needles designed to ionize sample components.

SHIELD leverages a customized large language model pipeline to detect and investigate sophisticated cyber threats with high accuracy and interpretability.

A novel transparent contact lens device enabling real-time monitoring of corneal curvature during electrochemical vision therapy.

Efficiently isolating specific biological components from complex mixtures is a cornerstone of modern biotechnology. UC Berkeley researchers have developed a robust method for the selective magnetic separation of target cells and molecules contained within microcompartments, allowing for the rapid isolation and recovery of high-purity biological samples. This approach is particularly effective for high-throughput screening and the analysis of rare cellular populations.

Preserving the functionality of biological molecules within synthetic environments remains a significant challenge in materials science. To address this, researchers at UC Berkeley have developed specialized plastic compositions designed to dramatically enhance the thermal stabilization of embedded proteins. These compositions utilize a strategic blend of salts and optional polymeric protectants to create a supportive microenvironment for the protein. By leveraging a synergistic effect between the salt and the polymer, the material prevents protein denaturation even when exposed to high temperatures that would typically lead to structural failure. This innovation allows for the creation of "living plastics" and bio-hybrid materials that maintain enzymatic or biological activity under demanding industrial or environmental conditions.

Advancing the field of robotic agility, this technology treats the complex challenge of bipedal balance and movement as a generative sequence problem. By framing physical movement similarly to language modeling, UC Berkeley researchers have developed a system where a humanoid robot predicts its next motor action as a "next token" based on a vast history of sensorimotor trajectories. The model is trained on diverse data, including real-world robotic walks and simulated movements, allowing it to anticipate the necessary joint adjustments and equilibrium shifts in real-time. This approach enables the robot to navigate uneven terrain and respond to external perturbations with a level of fluidity and adaptability that traditional, rigidly programmed control laws often struggle to achieve.

Bridging the gap between a language model’s next-word prediction and physical robot control, researchers at UC Berkeley have developed LLARVA (Large Language model for Robotic Vision and Action). This model utilizes a novel vision-action instruction tuning method that allows a robotic device to handle various tasks and environments without task-specific fine-tuning.